Experimental evidence exist which suggest that the pattern evoked electroretinogram (PERG) is generated by the ganglion cell layer of the retina. If this is the case then the PERG would be useful as a clinical tool to selectively test this level of the visual pathway. However, considerable experimental data and certain clinical observations conflict with the hypothesis that ganglion cells generate the PERG. The set of studies proposed here will examine the nature and origin of this response. Studies in the first part of this proposal are designed to replicate Maffei and Fiorentini's important cat PERG work. The flash and pattern evoked ERGs will be recorded before and at intervals after sectioning of the optic nerve. Such data are the basis for suggesting that the PERG is generated by the ganglion cells, but these studies have never been replicated by anyone other than the original investigators. In the second part of the proposal experiments similar to those of part one will be conducted on cats that are X ganglion cell deficient. These studies will provide evidence concerning the contribution of X and possibly Y cells to the PERG and the effects of subtotal damage to the optic nerve and ganglion cell population, as is the case in many ophthalmological problems. In the last part of this proposal the flash and pattern evoked ERGs will be examined before and after injection of retrogradely transported neurotoxins. These studies will directly test whether ganglion cells are the mechanisms that generate the PERG. Taken together these studies will not only provide information about the utility of the PERG as a clinical tool by telling us whether the ganglion cells specifically produce the PERG but also will provide information about the nature of the PERG and potentially facilitate our interpretation of the PERG in clinical situations.